Learning Interface Breakup: A Geometry-Conditioned Latent Surrogate for Spray Formation
📰 ArXiv cs.AI
Learn to predict spray formation using a geometry-conditioned latent surrogate, reducing the need for expensive high-fidelity simulations
Action Steps
- Implement a geometry-conditioned latent surrogate model to predict spray formation
- Use adaptive mesh refinement (AMR) to improve simulation accuracy
- Train the surrogate model on high-fidelity volume-of-fluid (VOF) simulations
- Evaluate the surrogate model's performance on unseen geometries and flow conditions
- Apply the surrogate model to iterative design exploration and optimization of spray nozzles
Who Needs to Know This
Researchers and engineers working on spray nozzle design and simulation can benefit from this approach to improve design exploration and optimization
Key Insight
💡 A geometry-conditioned latent surrogate can effectively predict spray formation, reducing the need for expensive high-fidelity simulations
Share This
🚀 Accelerate spray nozzle design with a geometry-conditioned latent surrogate 📈
Key Takeaways
Learn to predict spray formation using a geometry-conditioned latent surrogate, reducing the need for expensive high-fidelity simulations
Full Article
Title: Learning Interface Breakup: A Geometry-Conditioned Latent Surrogate for Spray Formation
Abstract:
arXiv:2606.16587v1 Announce Type: cross Abstract: Designing spray nozzles requires predicting how geometry shapes transient two-phase breakup, but high-fidelity volume-of-fluid (VOF) simulations with adaptive mesh refinement (AMR) are too expensive for iterative design exploration. Standard surrogate models are also challenged by this setting because both the liquid--gas interface and the underlying adaptive discretization evolve across time and geometries. We introduce a geometry-conditioned la
Abstract:
arXiv:2606.16587v1 Announce Type: cross Abstract: Designing spray nozzles requires predicting how geometry shapes transient two-phase breakup, but high-fidelity volume-of-fluid (VOF) simulations with adaptive mesh refinement (AMR) are too expensive for iterative design exploration. Standard surrogate models are also challenged by this setting because both the liquid--gas interface and the underlying adaptive discretization evolve across time and geometries. We introduce a geometry-conditioned la
DeepCamp AI